Employing a panel vector autoregression (PVAR) methodology, the paper addresses the nexus between military expenditures and two key macroeconomic variables, namely growth rates and investment spending using SIPRI’s new consistent time series dataset. Given data constraints, the sample chosen was a compromise between T and N for the empirical tests conducted. It consists of 65 countries and covers the period 1971–2014 that allows for a total of 2730 observations. The PVAR model is estimated for the entire sample as well as three income group sub-samples. Findings reported herein are not uniformed, and noteworthy differences between the three income groups were unearthed by the empirical tests conducted.